Toggling the mute button in Microsoft Lync for Mac via Applescript

Through liberal use of the dictionary for Microsoft Lync, the Accessibility Inspector, and a few web links, I’ve come up with this script which toggles the mute button in Lync for Mac.

tell application "Microsoft Lync"
end tell
tell application "System Events"
  tell process "Microsoft Lync"
    repeat with aWindow in (get every window)
      set aName to get the name of aWindow
      set initialName to ((characters 1 through 12 of aName) as string)
      if (initialName = "Conversation") then
        activate aWindow
        click checkbox 5 of splitter group 1 of aWindow
      end if
    end repeat
  end tell
end tell

I still want to hook it up to a hotkey, and ideally, have a different hotkey for mute and unmute, but that will have to be left for another session. I will hopefully update this post with more details as I figure out how to explain them.

rspec-rails 3 deprecation warnings scripts and fixes

Upon upgrading to rspec 2.99.0, I began to receive deprecation warnings for several things that had issues, and to do a few updates:

Because the project was using its syntax and stub_model I needed to add that functionality back into rspec in the :test group of my Gemfile:

  # Support for its syntax
  gem 'rspec-its', '~> 1.0.1’

  # Support for stubbing model in view specs:
  gem 'rspec-activemodel-mocks', '~> 1.0.1'

On top of that I had to do substitutions for be_true, be_false, and_return with block syntax, and change pending to skip for a few pending tests.

I was able to run the following find + sed commands for each of these (note that the .original part is Mac-specific):

find . -name '*.rb' -exec sed -i .original 's/be_true/be_truthy/g' {} +
find . -name '*.rb' -exec sed -i .original 's/be_false/be_falsey/g' {} +
find . -name '*.rb' -exec sed -i .original 's/and_return[ ]*{[ ]*\([^} ]*\)[ ]*}/and_return(\1)/g' {} +
find . -name '*.rb' -exec sed -i .original 's/^\([ ]*\)pending /skip /g' {} +

I haven’t completely foolproofed the last two regexes, so I’d make sure to have your work committed for each set of substitutions (and to make sure rspec still passes your tests, of course.)

And the cleanup:

find . -name '*original' -exec rm {} +

Lastly, to quiet the remaining deprecation warnings that I had, I added this to my spec_helper.rb:

Rspec.configure do |config|
# other config code
  config.mock_with :rspec do |mocks|
    mocks.yield_receiver_to_any_instance_implementation_blocks = true

# other config code

This configuration, as I understand it, will be standard with 3.0:

  mocks.yield_receiver_to_any_instance_implementation_blocks = true

This setting is required for 2.99 because you will unconditionally get a deprecation warning even if you set the type…


And I wanted to raise an error for any other deprecation warnings:


After all the dust had settled from 2.99, I upgraded rspec-rails to 3.0.0 and did a bundle update:

  gem 'rspec-rails', '~> 3.0.0' # Test framework

I’ve unearthed a few deprecation warnings about :should syntax, which I expect to update later.

One misleading deprecation warning:

Requiring `rspec/autorun` when running RSpec via the `rspec` command is deprecated. Called from /Users/tpowell/.rbenv/versions/2.1.1/lib/ruby/gems/2.1.0/gems/activesupport-4.1.1/lib/active_support/dependencies.rb:247:in `require'.

…was actually due to having require 'rspec/autorun' in my spec_helper.rb

mkdir with intermediate directories and rails generate view spec

I’ve been running into a lot of cases where I need to build intermediate directories for a path. Apparently, the correct option is -p

mkdir -p dir/tree/to/make

From the mkdir man page:

     -p      Create intermediate directories as required.  If this option is not specified, the full path prefix of each operand must already exist.  On the
             other hand, with this option specified, no error will be reported if a directory given as an operand already exists.  Intermediate directories
             are created with permission bits of rwxrwxrwx (0777) as modified by the current umask, plus write and search permission for the owner.

I’ve assigned it an alias in my .zlogin:

alias mkdirtree='mkdir -p'

Of course, this all came about because I needed to an generate rspec file for a view, which can be done with:

rails g rspec:view name/of/controller action another_action

Results in the following files in the spec/name/of/controller directory:


Preventing tinyMCE 4 from inserting <br /> within <code> blocks.


I was trying to hook in TinyMCE 4 as a editor that allows me to specify code blocks. However, in visual editing mode, <code> blocks ended up having embedded <br /> tags that displayed when rendered.

There is an old plugin, Pre Element Fix, that may have done what I wanted it to do, but it didn’t appear to be developed beyond TinyMCE 2.


It took a bit of work to handle spanning multiline with a match (/[^]*/) and hook into the editor setup (setup: function(editor) {...}).

Complete with initialization code, the following is what I came up with:

fixNewLines = function(content)
  var codeBlocks=content.match(/<code.*?>[^]*?<\/code>/m)

  if(codeBlocks == null) return content;
  for(var index=0; index < codeBlocks.length; index++) {
    content = content.replace(codeBlocks[index], codeBlocks[index].replace(/<br\s*\/?>/mgi, "\n"));
  return content;

    selector: "textarea",
    theme: "modern",
    entity_encoding : "raw",
    plugins: [
        "advlist autolink lists link image charmap print preview hr anchor pagebreak",
        "searchreplace wordcount visualblocks visualchars code fullscreen",
        "insertdatetime media nonbreaking save table contextmenu directionality",
        "emoticons template paste textcolor" ],
    toolbar1: "insertfile undo redo | styleselect | bold italic | alignleft aligncenter alignright alignjustify | bullist numlist outdent indent | link image",
    toolbar2: "print preview media | forecolor backcolor emoticons",
    image_advtab: true,
    templates: [
        {title: 'Test template 1', content: 'Test 1'},
        {title: 'Test template 2', content: 'Test 2'}
    setup: function(editor) {
      editor.on('BeforeSetContent', function(e) {
        e.content = fixNewLines(e.content);
      editor.on('SaveContent', function(e) {
        e.content = fixNewLines(e.content);


The key parts are the fixNewLines function, the setup hooks, the code plugin, and (possibly) the entity_encoding:"raw".

When you view the source code, you will still see break tags with the code sections. However, these will be cleared out on save.

Useful References That I Found

Syntax Highlighting Javascript

TinyMCE event research

Hook into onExecCommand event with tinymce 4

api:class.tinymce.Editor events

BeforeSetContent event.

SaveContent event.

            console.log('BeforeSetContent event', e);

Experiences so far with the 39″ Seiki 4k display as a monitor

I bought the 39″ Seiki 4k display on a whim as it was only $389 at Amazon and Newegg, after reading a post from another programmer about using it as a coding display. My initial findings and notes:

  • Make sure to have a snugly fitting Phillips head screwdriver before trying to mount the base.
  • Plugged work rMBP into it via the suppled(!) HDMI cable into the HDMI 1 input on the of the display.
  • There are only two scaling options for the rMBP: 3840×2160 (unscaled) and 1440×900 (not really worth the display if you’re going to do that.)
  • Dial sharpness down to 0 to be able to read your fonts.
  • Lower brightness below 50. I have it down to 24, but you may have a different preference.
  • Color was a bit too vivid…  I reduced to around 25.
  • I can divide my work window into quadrants and have full view of each:
    • Code window with 4 vertical and 2 horizontal split panes in MacVim.
    • iTerm window divided into 4 80×25 quadrants.
    • Browser window
    • Evernote window
  • The internal speakers on the TV sound like a cheap pair of speakers. More output than the rMBP internal speakers, but sound quality is a bit worse.
  • iTunes visualizer and YouTube videos don’t show the weakness in the 30Hz refresh rate, only a bit of higher speed mouse tracking has even remotely exposed it for me, but I’m don’t a desktop gamer.
  • Mouse speed vs being able to see your mouse pointer in motion can be a bit of a challenge. The larger display is more conducive to keyboard input for me.
  • Attempting to use the entire screen at once seems to introduce lag from eye-tracking, I’ve found it more useful to divide work areas on the screen. The benefit then comes from having everything on top at once.


I am now mostly using the settings listed in this article, except with a brightness of 35:

  • Picture Mode: User
  • Contrast: 38
  • Brightness: 45
  • Color: 36
  • Sharpness: 0
  • Color Temp: Warm

Filter chain halted as :find_page rendered or redirected in refinerycms

I ended up coming across this issue in Refinery CMS after switching databases.

I had forgotten that I had swapped out databases because I had moved on to changing some other configuration bits in following the RailsCast for Refinery CMS.

Fortunately, I found note on Refinery’s Getting Started page that deals with this:

If you get a 404 error on the homepage, try running the rake db:seed and then restarting the server.

rails generate paperclip_database:migration getting protected method `migration_file_name’ error

Quick fix (put this in application.rb or similar in order to run the rails generate paperclip_databasee:migration command):

module PaperclipDatabase
  module Generators
    class MigrationGenerator
      public :migration_file_name

There are two errors causing this error, one in paperclip and one in paperclip_database. Not sure what changes to Rails 4 caused this, but the issue is fixed in this commit in paperclip, which probably needs to be mirrored in paperclip_database. (Even if I point to the github repo for paperclip, I still get the error due to protected being specified on paperclip_database as well.

The error only appears to affect the generator.

If you’d rather not have a temporary monkey patch, you can write the migration by hand:

Example (for a User with an avatar attachment):

class CreateUserAvatar < ActiveRecord::Migration
  def self.up
    create_table :avatars do |t|
      t.integer    :user_id
      t.string     :style
      t.binary     :file_contents

  def self.down
    drop_table :avatars

Bundler gave the error “Could not find {some gem} in any of the sources”, processing Gemfile in wrong directory


I was persistently getting the error:

Bundler gave the error "Could not find json-1.8.1 in any of the sources" while processing "/Users/tpowell/projects/some_other_project/Gemfile". Perhaps you forgot to run "bundle install"?

While trying to fix a bug in paperclip_database that was similar to Paperclip Gerenator – No Method Error in paperclip

I eventually found an issue referencing this problem in the rbenv github repo.:

It is recommended to not use the rbenv-bundler plugin, as it has been the source of many problems


I ended up uninstalling the rbenv-bundler plugin as a result.

But hey, kudos to rbenv-bundler author for taking the complaint in stride.

CoffeeScript event handlers only working after page refresh

I was able to find a resolution on StackOverflow to fix an event handler.

I initially opted to added $(document).ready(...) and $(document).on('page:load', ...) handlers as follows:

ready = ->
  $('form').on 'click', '.clickable_class', (event) ->
    # doStuff

# added lines:
$(document).on('page:load', ready)

However, another answer recommended adding the jquery-turbolinks instead. Browsing the source code, it appears that this gem effectively auto-registers those handlers instead.

After installing jquery-turbolinks, I was able to add require jquery.turbolinks to my app/assets/javascripts/application.js manifest:

//= require jquery
//= require jquery.turbolinks
//= require jquery_ujs
//= require turbolinks
//= require_tree .

With that, my script could go back to:

$ ->
  $('form').on 'click', '.clickable_class', (event) ->
    # doStuff

# added lines:
$(document).on('page:load', ready)

ActionDispatch::Cookies::CookieOverflow and Rails 4.1

Ran into this error when adding a new route.

Turns out that I was stuffing the model.errors into the alert flash, making the session cookie relatively huge.

While stuffing all that data into the cookie wasn’t intended, it made me notice that we were storing the session in a cookie instead of a database like with other projects. It turns out that the ActiveRecord session store has been extracted out of Rails into its own gem recently, the details of which are available on the activerecord-session_store GitHub project page.